1. Field of the Invention
The present invention relates to techniques for assisting on image-based diagnosis using plural virtual endoscopic images generated from a three-dimensional image.
2. Description of the Related Art
In recent years, diagnosis was performed in various clinical scenes by observing a three-dimensional image obtained by imaging a patient. Especially, for a tubular structural organ, such as a large intestine and a bronchus or bronchi, which is observable by using an endoscope, a method for observing a region of interest, such as a lesion, has been drawing attention. In the method, plural images resembling endoscopic images (hereinafter, referred to as virtual endoscopic images), and which are virtually imaged from the inside of the tubular structure, are generated along the tubular structure from a CT image, and the virtual endoscopic images of plural positions at plural angles are successively displayed. Such image-based diagnosis using virtual endoscopic images can reduce a burden on the patient in examination, because it is not necessary to actually insert an endoscope into the patient. Further, even if an obstructed region exists in the tubular structure and the endoscope is not insertable, it is possible to display, based on the CT image, the condition of a part of the tubular structure behind the obstructed region. Therefore, there are growing expectations for the image-based diagnosis using such virtual endoscopic images.
Japanese Unexamined Patent Publication No. 2000-135215 (Patent Document 1) proposes a method in which a three-dimensional image is generated for a tube, such as a bronchus or bronchi, and a path to a target point along the tube is obtained in advance in the three-dimensional image, and a virtual endoscopic image along the path is used, as a guide image, in actual observation using an endoscope. Further, PCT Japanese Publication No. 2008-529578 (Patent Document 2) proposes a method in which a trigger is stored in such a manner to be related to a region of interest of a user. When virtual endoscopic images are successively displayed along a path as if a motion image is displayed, enhanced display is possible for the region in which the trigger has been set. A resolution, a display rate, a display time, the magnification of an image, a view angle and the like are changed based on a set viewing protocol.